Re: BUG #4319: lower()/upper() does not know about UNICODE case mapping

On Jul 22, 11:53 am, pete(dot)(dot)(dot)(at)gmx(dot)net (Peter Eisentraut) wrote:
> Am Tuesday, 22. July 2008 schrieb valgog:
>
> > Why Postgres allows creating UNICODE database with the locale, that
> > can possibly corrupt my data?
>
> It doesn't allow it, as of 8.3.  In 8.2 it does, but we have fixed that, for
> the reasons that are becoming obvious to you now.
>
> Perhaps part of the problem is that en_EN isn't actually a valid locale, as
> far as I can tell, unless SUSE has invented a new country. :)  Try locale -a
> and pick one from that list.
>
> --
> Sent via pgsql-bugs mailing list (pgsql-b(dot)(dot)(dot)(at)postgresql(dot)org)
> To make changes to your subscription:http://www.postgresql.org/mailpref/pgsql-bugs

Ok, I have checked in databases with locale set to one of the utf8
locales convert data correctly. But I still do not understand, why
postgres initdb allows using non-UTF8 locales together with UNICODE
database encoding.

Another question. Is it possible to change the current Database
LC_CTYPE on the database without recreating it with initdb and
reimporting all the data. I would rather prefer my indexes recreated,
rather then reimporting database (cannot afford such a long time out
of service :(